A power distribution network actual effect data driven line self-healing control method and system

Abstract

A kind of power distribution network actual effect data driven line self-healing control method and system, by real-time acquisition power distribution network actual effect operation data, utilize graph traversal algorithm to generate real-time network topology structure, in combination with key node electrical quantity data and fault indicator alarm information accurately locate fault section;Based on multi-objective optimization algorithm, the model with switch operation number minimization, power supply recovery rate maximization as target is constructed, and the optimal self-healing strategy is generated;After dynamic simulation is completed, safety check is carried out, and execution effect is checked in real time, and the perception-decision-check-execution-feedback closed loop is formed.The present application solves the problems of inaccurate positioning, insufficient risk prediction, one-sided optimization and poor self-adaptation in the prior art through dynamic topology analysis, multi-objective collaborative optimization, prospective safety check and self-adaptive fault-tolerant mechanism, and realizes the intelligent self-healing control of power distribution network safety, economy and high efficiency.

Description

Technical Field

[0001] This invention belongs to the field of smart distribution network automation technology, and in particular relates to a line self-healing control method and system driven by real-time data of distribution networks. Background Technology

[0002] With the increasing demands for power supply reliability, self-healing control technology for distribution networks has become the core of smart distribution networks. As a crucial link in power transmission, the rapid self-healing and recovery of the distribution network after a fault directly impacts residents' lives, enterprise production, and the stable operation of critical facilities. Existing self-healing control methods largely rely on centralized computing at the master station, failing to deeply integrate real-time data for dynamic optimization, and thus cannot meet the needs of safe, efficient, and economical self-healing in distribution networks. Existing self-healing control methods, which rely heavily on centralized computing at the master station, have the following drawbacks: (1) Rigid fault location mechanism: Existing methods, such as patent document 1 with publication number CN104300539A, achieve accurate location of the fault area by detecting voltage pulsation signals (such as 3U0 when a single-phase grounding occurs). However, this existing technology fails to deeply associate fault information with the real-time changing network topology, and only performs simple signal superposition, making the location accuracy susceptible to interference; (2) Lack of strategy security verification: Existing technologies, such as patent document 2 with publication number CN111049112A, collect data such as current and switch status in real time through a multi-functional power monitoring terminal, and automatically verify whether the voltage and current meet the constraints after fault isolation, ensuring that the recovery strategy will not cause overload or voltage over-limit; at the same time, the node network variable logic connection technology is adopted, and the field terminal equipment can cooperate with each other, and can still achieve fault isolation and recovery through local decision-making when communication is interrupted. However, the recovery strategy generated by this method lacks forward-looking safety simulation verification and cannot predict whether the strategy will cause secondary risks such as line overload and voltage over-limit after execution; (3) Single optimization target: When generating self-healing strategy, it often only aims to maximize the recovery of power supply, ignoring the comprehensive optimization of multiple actual operation indicators such as the number of switching operations and line load, resulting in the generated strategy being able to restore power supply but possibly complex to operate, poor in economy or not meeting the long-term optimal operation; (4) Insufficient system adaptive capability: In complex scenarios such as limited recovery capacity or switch failure, Patent Document 2 lacks dynamic adjustment and priority guarantee mechanism, and cannot achieve intelligent fault tolerance and optimized recovery.

[0003] Therefore, there is an urgent need for a self-healing control method for distribution networks that can deeply integrate real-time data, achieve accurate fault location, and perform multi-objective collaborative optimization and forward-looking safety verification. Summary of the Invention

[0004] The purpose of this application is to provide a data-driven line self-healing control method and system for distribution networks. By constructing unified, high-quality real-time operational data and performing dynamic topology analysis, a realistic and reliable power grid state foundation is provided for fault location and self-healing strategy generation, significantly improving the accuracy of decision-making and response speed. Simultaneously, by employing a multi-objective optimization algorithm, under strict constraints on line load, multiple indicators such as the number of switch operations and power restoration rate are comprehensively weighed. This overcomes the limitation of traditional methods that only pursue the maximum restored power supply, automatically generating a strategy that achieves the best balance between operational risk, restoration efficiency, and operational economy. This upgrades self-healing control from "restoring power supply" to "safe, economical, and efficient optimal restoration." Furthermore, by introducing dynamic simulation safety verification before execution, risks such as line overload and voltage exceeding limits can be proactively identified and avoided, preventing secondary faults caused by blind operation and enhancing the system's safety and reliability. Combined with execution feedback and load priority management, the system possesses a high degree of intelligence and adaptability to cope with complex scenarios such as switch refusal and limited power supply capacity, ultimately forming a complete technical closed loop of "perception-decision-verification-execution-feedback" to achieve intelligent self-healing control.

[0005] The present invention adopts the following technical solution.

[0006] A data-driven line self-healing control method for distribution networks includes: Real-time acquisition of effective operational data of the power distribution network, and generation of real-time network topology based on the operational data; Real-time monitoring of electrical quantity data of the power root node in the real-time network topology, and fault determination based on the electrical quantity data; After determining that a fault has occurred, the faulty section is located by combining the alarm status information of the fault indicator and the real-time network topology. A multi-objective optimization model is constructed with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit. Based on the multi-objective optimization model, an optimal self-healing strategy for isolating the faulty sections and restoring power supply to the non-faulty sections is generated. The optimal self-healing strategy is simulated and its security is verified through dynamic simulation. If the security check passes, the optimal self-healing strategy will be deployed and executed, and the execution effect will be verified in real time.

[0007] More preferably, the real-time acquisition of effective operating data of the distribution network and the generation of a real-time network topology based on the operating data specifically includes the following steps: The operational data is standardized and its quality is verified to obtain the processed operational data. Construct the electrical wiring diagram of the power distribution network, and based on the electrical wiring diagram, construct an initial global topology structure with power distribution equipment as vertices and electrical connections between equipment as edges; Extract the real-time opening / closing status data of all switches from the processed operating data, map it onto the initial global topology, and generate a dynamic topology. Using all power supply points as root nodes, the dynamic topology is traversed to generate a real-time network topology containing information on each connected subgraph and power supply path.

[0008] More preferably, the real-time monitoring of electrical quantity data of the power root node in the real-time network topology, and the fault determination based on the electrical quantity data, specifically includes the following steps: The electrical quantity data includes the three-phase current, zero-sequence current, and voltage at the power source root node; If any phase current exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting, and the duration exceeds the preset time threshold, and the voltage value is lower than the preset voltage threshold, then a fault is determined to have occurred in the distribution network.

[0009] More preferably, the specific steps for locating the faulty section by combining the alarm status information of the fault indicator and the real-time network topology include: Obtain the alarm status information and corresponding ID of the fault indicator; Based on the ID, the alarm status information is mapped to the corresponding position in the real-time network topology, and the connected subgraph to which it belongs is queried according to the information of the position; Based on the power supply path information corresponding to the connected subgraph, the alarm status of each fault indicator on the path is queried sequentially. The section between the last fault indicator that triggered an alarm and the first fault indicator that did not trigger an alarm is defined as the fault section.

[0010] More preferably, the method for generating the optimal self-healing strategy for isolating the faulty section and restoring power supply to the non-faulty section based on the multi-objective optimization algorithm specifically includes: Solving the multi-objective optimization model yields a Pareto front solution set containing multiple candidate self-healing strategies; A comprehensive evaluation function is constructed based on the power restoration rate and the total number of switching operations. The comprehensive evaluation function is then used to calculate the comprehensive evaluation value corresponding to each candidate self-healing strategy. The candidate self-healing strategy with the highest comprehensive evaluation value is selected as the optimal self-healing strategy.

[0011] More preferably, the specific steps of performing dynamic simulation security verification on the optimal self-healing strategy include: Obtain the physical parameters of the distribution network, and construct a dynamic simulation model of the distribution network based on the processed operating data and the physical parameters of the distribution network. The reconstructed network structure data and switching operation sequence in the optimal self-healing strategy are input into the dynamic simulation model to obtain simulation results. The simulated electrical quantities are extracted from the simulation results, including the voltage values ​​of each node and the short-circuit current values ​​of each line; Calculate the deviation between the voltage value of each node and the preset rated voltage, and determine whether the deviation is less than the preset deviation threshold. Determine whether the short-circuit current value of each line is less than the rated breaking current value of its corresponding circuit breaker; If both judgments are yes, the verification passes and the optimal self-healing strategy is executed. Otherwise, the over-limit location is determined based on the simulation results, and the deviation value, short-circuit current value and their corresponding over-limit location data are input into a preset constraint condition lookup table to obtain the corresponding hard constraint conditions; The hard constraints are fed back to the multi-objective optimization model, and the corrected optimal self-healing strategy is generated by solving the problem again.

[0012] More preferably, if the security check passes, the optimal self-healing strategy is deployed and executed, and the execution effect is verified in real time. Specific steps include: Within a preset time period, determine whether a corresponding switch change confirmation signal has been received; if so, the execution is considered successful. If not, the switch that failed to execute is marked as inoperable, and the inoperable state is fed back to the multi-objective optimization model as a new constraint to regenerate the optimal self-healing strategy after excluding this switch.

[0013] More preferably, the step of marking the failed switch as an inoperable state and feeding the inoperable state back to the multi-objective optimization model as a new constraint includes the following steps: Add the switch identifier that failed to execute to the set of inoperable switches; When resolving the multi-objective optimization model, the decision variable corresponding to the switch in the set is fixed to the current state and removed from the search space of the optimization variables; The search space is re-optimized to generate the optimal self-healing strategy after excluding all inoperable switches.

[0014] More preferably, it also includes: If, after implementing the optimal self-healing strategy, the power supply recovery rate of the non-faulty section is less than the preset power supply recovery rate threshold; Then obtain the predefined priority labels of each load node, including primary load, secondary load and tertiary load; Based on the priority labels, set corresponding power recovery rate constraints for load nodes with different priorities; The minimum power recovery rate constraint is integrated as a new hard constraint into the multi-objective optimization model, and the modified optimal self-healing strategy is generated by solving the model again.

[0015] This invention also proposes a data-driven line self-healing control system for distribution networks, comprising a data acquisition module, a fault diagnosis module, a fault location module, a self-healing strategy generation module, a strategy security verification module, and an execution module. The data acquisition module collects real-time operational data of the power distribution network and generates a real-time network topology based on the operational data. The fault diagnosis module monitors the electrical quantity data of the power root node in the real-time network topology and diagnoses faults based on the electrical quantity data. The fault location module, after determining that a fault has occurred, locates the faulty segment by combining the alarm status information of the fault indicator and the real-time network topology. The self-healing strategy generation module constructs a multi-objective optimization model with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit; based on the multi-objective optimization model, it generates the optimal self-healing strategy for isolating the faulty section and restoring the power supply to the non-faulty section. The strategy security verification module performs simulated security verification on the optimal self-healing strategy through dynamic simulation. If the security check passes, the execution module will issue the optimal self-healing strategy for execution and verify the execution effect in real time.

[0016] The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of Embodiment 1.

[0017] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the embodiments.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention provides a real and reliable power grid state basis for fault location and self-healing strategy generation by constructing unified, high-quality, and effective operational data and performing dynamic topology analysis, which significantly improves the accuracy of decision-making and response speed. 2. This invention, by employing a multi-objective optimization algorithm, comprehensively weighs multiple indicators such as the number of switching operations and power restoration rate under strict constraints on line load. It overcomes the limitation of traditional methods that only pursue the maximum amount of restored power. It can automatically generate a strategy that achieves the best balance between operational risk, restoration efficiency, and operational economy, thus upgrading self-healing control from "restoring power" to "safe, economical, and efficient optimal restoration". 3. By introducing dynamic simulation safety verification before execution, this invention can proactively identify and avoid risks such as line overload and voltage exceeding limits, eliminate secondary faults caused by blind operation, and enhance the safety and reliability of the system. At the same time, combined with execution feedback and load priority management, the system has a high degree of intelligence and adaptability to cope with complex scenarios such as switch failure and limited power supply capacity, and finally forms a complete technical closed loop of "perception-decision-verification-execution-feedback" to achieve intelligent self-healing control. Attached Figure Description

[0019] Figure 1 This is a flowchart of a line self-healing control method driven by real-time data of a power distribution network according to the present invention. Figure 2 This is a flowchart of the line self-healing control method driven by real-time data of the distribution network provided in the embodiments of the present invention; Figure 3 This is a flowchart illustrating the generation of real-time network topology in the line self-healing control method driven by real-time data of the distribution network provided in this embodiment of the invention. Figure 4 This is a flowchart illustrating the location of fault sections in the line self-healing control method driven by real-time data of the distribution network provided in this embodiment of the invention. Figure 5 This is a flowchart illustrating the generation of the optimal self-healing strategy for the line self-healing control method driven by real-time data of the distribution network provided in this embodiment of the invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0021] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] The present invention proposes the following technical solution: This invention proposes a line self-healing control method driven by real-time data in a distribution network, such as... Figure 1 As shown, it includes the following steps: Real-time acquisition of effective operational data of the power distribution network, and generation of real-time network topology based on the operational data; The operational data is standardized and its quality is verified to obtain the processed operational data. Construct the electrical wiring diagram of the power distribution network, and based on the electrical wiring diagram, construct an initial global topology structure with power distribution equipment as vertices and electrical connections between equipment as edges; Extract the real-time opening / closing status data of all switches from the processed operating data, map it onto the initial global topology, and generate a dynamic topology. Using all power supply points as root nodes, the dynamic topology is traversed to generate a real-time network topology containing information on each connected subgraph and power supply path.

[0023] Real-time monitoring of electrical quantity data of the power root node in the real-time network topology, and fault determination based on the electrical quantity data; The electrical quantity data includes the three-phase current, zero-sequence current, and voltage at the power source root node; If any phase current exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting, and the duration exceeds the preset time threshold, and the voltage value is lower than the preset voltage threshold, then a fault is determined to have occurred in the distribution network.

[0024] After determining that a fault has occurred, the faulty section is located by combining the alarm status information of the fault indicator and the real-time network topology. Obtain the alarm status information and corresponding ID of the fault indicator; Based on the ID, the alarm status information is mapped to the corresponding position in the real-time network topology, and the connected subgraph to which it belongs is queried according to the information of the position; Based on the power supply path information corresponding to the connected subgraph, the alarm status of each fault indicator on the path is queried sequentially. The section between the last fault indicator that triggered an alarm and the first fault indicator that did not trigger an alarm is defined as the fault section.

[0025] A multi-objective optimization model is constructed with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit. Based on the multi-objective optimization model, an optimal self-healing strategy for isolating the faulty sections and restoring power supply to the non-faulty sections is generated. Solving the multi-objective optimization model yields a Pareto front solution set containing multiple candidate self-healing strategies; A comprehensive evaluation function is constructed based on the power restoration rate and the total number of switching operations. The comprehensive evaluation function is then used to calculate the comprehensive evaluation value corresponding to each candidate self-healing strategy. The candidate self-healing strategy with the highest comprehensive evaluation value is selected as the optimal self-healing strategy.

[0026] The optimal self-healing strategy is simulated and its security is verified through dynamic simulation. Obtain the physical parameters of the distribution network, and construct a dynamic simulation model of the distribution network based on the processed operating data and the physical parameters of the distribution network. The reconstructed network structure data and switching operation sequence in the optimal self-healing strategy are input into the dynamic simulation model to obtain simulation results. The simulated electrical quantities are extracted from the simulation results, including the voltage values ​​of each node and the short-circuit current values ​​of each line; Calculate the deviation between the voltage value of each node and the preset rated voltage, and determine whether the deviation is less than the preset deviation threshold. Determine whether the short-circuit current value of each line is less than the rated breaking current value of its corresponding circuit breaker; If both judgments are yes, the verification passes and the optimal self-healing strategy is executed. Otherwise, the over-limit location is determined based on the simulation results, and the deviation value, short-circuit current value and their corresponding over-limit location data are input into a preset constraint condition lookup table to obtain the corresponding hard constraint conditions; The hard constraints are fed back to the multi-objective optimization model, and the corrected optimal self-healing strategy is generated by solving the problem again.

[0027] If the security check passes, the optimal self-healing strategy will be deployed and executed, and the execution effect will be verified in real time.

[0028] Within a preset time period, determine whether a corresponding switch change confirmation signal has been received; if so, the execution is considered successful. If not, the switch that failed to execute is marked as inoperable, and the inoperable state is fed back to the multi-objective optimization model as a new constraint to regenerate the optimal self-healing strategy after excluding this switch.

[0029] The steps of marking the failed switch as inoperable and feeding the inoperable state back to the multi-objective optimization model as a new constraint include: Add the switch identifier that failed to execute to the set of inoperable switches; When resolving the multi-objective optimization model, the decision variable corresponding to the switch in the set is fixed to the current state and removed from the search space of the optimization variables; The search space is re-optimized to generate the optimal self-healing strategy after excluding all inoperable switches.

[0030] Also includes: If, after implementing the optimal self-healing strategy, the power supply recovery rate of the non-faulty section is less than the preset power supply recovery rate threshold; Then obtain the predefined priority labels of each load node, including primary load, secondary load and tertiary load; Based on the priority labels, set corresponding power recovery rate constraints for load nodes with different priorities; The minimum power recovery rate constraint is integrated as a new hard constraint into the multi-objective optimization model, and the modified optimal self-healing strategy is generated by solving the model again.

[0031] This invention also proposes a data-driven line self-healing control system for distribution networks, comprising a data acquisition module, a fault diagnosis module, a fault location module, a self-healing strategy generation module, a strategy security verification module, and an execution module. The data acquisition module collects real-time operational data of the power distribution network and generates a real-time network topology based on the operational data. The fault diagnosis module monitors the electrical quantity data of the power root node in the real-time network topology and diagnoses faults based on the electrical quantity data. The fault location module, after determining that a fault has occurred, locates the faulty segment by combining the alarm status information of the fault indicator and the real-time network topology. The self-healing strategy generation module constructs a multi-objective optimization model with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit; based on the multi-objective optimization model, it generates the optimal self-healing strategy for isolating the faulty section and restoring the power supply to the non-faulty section. The strategy security verification module performs simulated security verification on the optimal self-healing strategy through dynamic simulation. If the security check passes, the execution module will issue the optimal self-healing strategy for execution and verify the execution effect in real time.

[0032] The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of Embodiment 1.

[0033] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the embodiments.

[0034] Example 1 Please refer to Figure 2 , Figure 2 This is a flowchart of a line self-healing control method driven by real-time data of a distribution network, as described in some embodiments of this application. This line self-healing control method driven by real-time data of a distribution network is used in terminal equipment, such as computers and mobile terminals. The line self-healing control method driven by real-time data of a distribution network includes the following steps: S11. Collect real-time operational data of the power distribution network and generate real-time network topology using a graph traversal algorithm; S12. Monitor the electrical quantity data of key nodes in the real-time network topology and determine faults based on the electrical quantity data. S13. Locate the faulty section by combining the alarm status information of the fault indicator and the real-time network topology. S14. Generate the optimal self-healing strategy for isolating faulty sections and restoring power supply to non-faulty sections based on a multi-objective optimization algorithm; S15. Perform dynamic simulation to verify the safety of the optimal self-healing strategy. S16. If the security check passes, the optimal self-healing strategy will be deployed and executed, and the execution effect will be verified in real time.

[0035] It should be noted that by constructing unified, high-quality, and effective operational data and performing dynamic topology analysis, a realistic and reliable power grid state foundation is provided for fault location and self-healing strategy generation, significantly improving the accuracy of decision-making and response speed. Simultaneously, leveraging multi-objective optimization algorithms, under strict constraints on line load, a comprehensive balance is struck between multiple indicators such as the number of switching operations and power restoration rate. This overcomes the limitations of traditional methods that only pursue maximum power restoration, automatically generating strategies that achieve the optimal balance between operational risk, restoration efficiency, and operational economy. This upgrades self-healing control from simply "restoring power" to "safe, economical, and efficient optimal restoration." Furthermore, by introducing dynamic simulation safety verification before execution, risks such as line overload and voltage exceeding limits can be proactively identified and avoided, preventing secondary faults caused by blind operations and enhancing system safety and reliability. Combined with execution feedback and load priority management, the system possesses a high degree of intelligence and adaptability to cope with complex scenarios such as switch failure and limited power supply capacity, ultimately forming a complete technical closed loop of "perception-decision-verification-execution-feedback" to achieve intelligent self-healing control.

[0036] Please refer to Figure 3 , Figure 3This is a flowchart illustrating the generation of a real-time network topology using a distribution network real-time data-driven line self-healing control method, as described in some embodiments of this application. According to embodiments of the present invention, the real-time acquisition of distribution network real-time operating data and the generation of a real-time network topology using a graph traversal algorithm include: S21. Real-time collection of effective operation data of the power distribution network and standardization processing and data quality verification; S22. Based on the electrical wiring diagram of the power distribution network, construct an initial global topology with power distribution equipment as vertices and electrical connections between equipment as edges; S23. Extract the real-time opening / closing status data of all switches from the standardized and quality-verified operating data, map it onto the initial global topology, and generate a dynamic topology. S24. Using all power supply points as root nodes, the dynamic topology is traversed using a preset breadth-first search algorithm or other traversal algorithm to generate a real-time network topology containing information about each connected subgraph and its power supply path.

[0037] It should be noted that the raw operational data collected from the distribution terminals is cleaned and organized, and its completeness and rationality are verified. Abnormal data is identified and processed to ensure data quality and provide a reliable foundation for subsequent analysis. Then, various power distribution equipment (including circuit breakers, load switches, sectionalizing switches, fault indicators, transformers, etc.) are abstracted as vertices of a graph, and the electrical connection segments between equipment are abstracted as edges of the graph, generating an initial global topology. The real-time status of all switches (including circuit breakers, load switches, sectionalizing switches, etc.) is extracted from the preprocessed operational data and "mapped" onto the initial global topology (e.g., closing switches correspond to "edge connected," and opening switches correspond to "edge disconnected"), generating a "dynamic topology" that reflects the current actual connection relationships. The real-time network topology includes the current grid connectivity partitions and power supply path information: each connected subgraph represents an area independently powered by a single power source, and the power supply path information includes the electrical connection relationships and power flow direction from the power source point to each load node within the area.

[0038] According to an embodiment of the present invention, the real-time monitoring of electrical quantity data of key nodes in the real-time network topology, and the fault determination based on the electrical quantity data, includes: Real-time acquisition of three-phase current, zero-sequence current, and voltage of the power source root node in the real-time network topology; If any phase current exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting, and the duration exceeds the preset time threshold, and the voltage value is lower than the preset voltage threshold, then a fault is determined to have occurred in the distribution network.

[0039] It should be noted that, in order to achieve accurate fault identification and avoid misjudgment or omission, the power source root node is selected as the key monitoring node (because changes in the electrical quantities of the power source node can reflect the state of the entire connected sub-diagram). The three-phase current, zero-sequence current and voltage are collected in real time. When making a judgment, three conditions must be met simultaneously: the current of any phase exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting (covering overcurrent and grounding faults), the abnormal duration exceeds the preset time threshold (excluding transient interference), and the voltage value is lower than the preset voltage threshold (further confirming the fault rather than load fluctuation). If all three conditions are met, the distribution network is judged to have a fault.

[0040] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating the location of fault sections in a distribution network real-time data-driven line self-healing control method according to some embodiments of this application. According to an embodiment of the present invention, the location of fault sections by combining alarm status information from fault indicators and real-time network topology includes: S31. Obtain the fault indicator alarm status information and its corresponding ID; S32. Map the alarm status information of the fault indicator to the corresponding position in the real-time network topology according to the fault indicator ID, and query the connected subgraph to which it belongs based on the position information; S33. Based on the power supply path information corresponding to the connected subgraph, query the alarm status of each fault indicator on the path in sequence. S34. The section between the last fault indicator that triggered an alarm and the first fault indicator that did not trigger an alarm is determined to be a fault section.

[0041] It should be noted that, assuming that the system does indeed have a fault, the fault indicator ID is used to map the alarm status information (alarm or no alarm) to the corresponding location in the real-time network topology in order to accurately locate the smallest range (segment) where the fault occurred.

[0042] Please refer to Figure 5 , Figure 5 This is a flowchart illustrating the generation of the optimal self-healing strategy for a distribution network-driven line self-healing control method in some embodiments of this application. According to an embodiment of the present invention, the generation of the optimal self-healing strategy for isolating faulty sections and restoring power supply to non-faulty sections based on a multi-objective optimization algorithm includes: S41. Construct a multi-objective optimization model with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit. S42. Use a pre-defined multi-objective genetic algorithm or other solution methods such as multi-objective particle swarm optimization algorithm, multi-objective differential evolution algorithm, etc. to solve the multi-objective optimization model and obtain a Pareto front solution set containing multiple candidate self-healing strategies; S43. Construct a comprehensive evaluation function based on the power supply recovery rate and the total number of switching operations, and use the comprehensive evaluation function to calculate the comprehensive evaluation value corresponding to each candidate self-healing strategy. S44. Select the candidate self-healing strategy with the highest comprehensive evaluation value as the optimal self-healing strategy.

[0043] It should be noted that the control parameters such as population size, number of iterations, crossover probability, and mutation probability in the multi-objective genetic algorithm are determined based on the performance obtained from actual implementation. By leveraging the multi-objective optimization algorithm, under strict constraints on line load, multiple indicators such as the number of switching operations and power restoration rate are comprehensively weighed. This overcomes the limitation of traditional methods that only pursue the maximum amount of restored power, and can automatically generate a strategy that achieves the best balance between operational risk, restoration efficiency, and operational economy. This upgrades self-healing control from "restoring power" to "safe, economical, and efficient optimal restoration." The power restoration rate refers to the ratio of the total load that successfully restores power after the self-healing strategy is executed to the total load of the non-faulty section that lost power due to the fault. In this embodiment, the comprehensive evaluation function is obtained by multiplying the power restoration rate and the total number of switching operations by weight coefficients and then taking the difference. Each candidate self-healing strategy includes a defined set of switching operations, an operation sequence, and a reconstructed network structure. The weight coefficients are determined based on actual experimental performance.

[0044] According to an embodiment of the present invention, the step of performing simulated security verification of the optimal self-healing strategy through dynamic simulation includes: Obtain the physical parameters of the distribution network, and construct a dynamic simulation model of the distribution network based on real-time operating data and the physical parameters of the distribution network; The reconstructed network structure data and switching operation sequence in the optimal self-healing strategy are input into the dynamic simulation model to obtain simulation results. The simulated electrical quantities are extracted from the simulation results, including the voltage values ​​of each node and the short-circuit current values ​​of each line. Calculate the deviation between the voltage value of each node and the preset rated voltage, and determine whether the deviation is less than the preset deviation threshold. Determine whether the short-circuit current value of each line is less than the rated breaking current value of its corresponding circuit breaker; If both judgments are yes, the verification passes and the optimal self-healing strategy is executed. Otherwise, the over-limit location is determined based on the simulation results, and the deviation value, short-circuit current value and their corresponding over-limit location data are input into a preset constraint condition lookup table to obtain the corresponding hard constraint conditions; The hard constraints are fed back to the multi-objective optimization model, and the corrected optimal self-healing strategy is generated by solving the problem again.

[0045] It should be noted that when performing dynamic simulation to verify the optimal self-healing strategy for safety, if the node voltage deviation exceeds the preset deviation threshold or the line short-circuit current exceeds the rated breaking current of the corresponding circuit breaker, it indicates that the self-healing strategy violates safety operation constraints. In this case, the simulation results need to be analyzed to identify the specific violations, and the hard constraints should be determined through a preset constraint reference table. For example, the simulation results may determine whether the short-circuit current of a certain line exceeds the rated breaking current of its terminal circuit breaker, or whether the deviation of the node voltage from the rated voltage in a certain area exceeds the allowable range. The hard constraints (such as "the short-circuit current of line A shall not exceed 12kA" or "the voltage deviation of node B shall be controlled within ±5%) are obtained by looking up the constraint reference table and directly fed back into the multi-objective optimization model. When the model is re-solved, these new constraints will be included in the optimization boundary along with the original line load rate constraints, thereby generating a modified optimal self-healing strategy that can avoid such safety risks, ensuring that the strategy meets both the power supply restoration objective and the requirements for safe operation of the distribution network.

[0046] According to an embodiment of the present invention, if the security verification passes, the optimal self-healing strategy is issued and executed, and the execution effect is verified in real time, including: Within a preset time period, determine whether a corresponding switch change confirmation signal has been received; if so, the execution is considered successful. If not, the switch that failed to execute is marked as inoperable, and this state is fed back to the multi-objective optimization model as a new constraint to regenerate the optimal self-healing strategy after excluding this switch.

[0047] It should be noted that if the safety check passes, the optimal self-healing strategy will be issued and executed, and the execution effect will be verified in real time. The focus is on solving the fault tolerance problem of switch operation failure. After the strategy is issued, the switch change confirmation signal will be monitored within a preset time period. If the signal is received, the operation is determined to be successful. If no signal is received, the switch is marked as inoperable. This state is fed back to the multi-objective optimization model as a new constraint, and the optimal self-healing strategy to exclude the switch is regenerated to avoid the interruption of the self-healing process due to a single switch failure.

[0048] Switches that fail to execute are marked as inoperable, and this state is fed back as a new constraint to the multi-objective optimization model, specifically including: Add the switch identifier that failed to execute to the set of inoperable switches; When resolving the multi-objective optimization model, the decision variable corresponding to the switch in the set is fixed to the current state and removed from the search space of the optimization variables; The search space is re-optimized to generate the optimal self-healing strategy after excluding all inoperable switches.

[0049] According to an embodiment of the present invention, it further includes: If the power supply recovery rate of the non-faulty section is less than the preset power supply recovery rate threshold after the optimal self-healing strategy is implemented; Obtain the predefined priority labels for each load node from the distribution network management system, including primary load, secondary load, and tertiary load; Based on the priority labels, set corresponding power recovery rate constraints for load nodes with different priorities; The minimum power restoration rate constraint is integrated as a new hard constraint into the multi-objective optimization model, and the modified optimal self-healing strategy is generated by solving the model again.

[0050] It should be noted that load priority labels are defined in the distribution network dispatching system, dividing all load nodes into three levels: Level 1 loads (critical loads): such as hospital intensive care units, city traffic lights, and emergency command centers, which require priority power restoration; Level 2 loads (important loads): such as residential communities and commercial complexes; Level 3 loads (general loads): such as industrial auxiliary equipment and non-essential commercial facilities. Priority is given to ensuring a restoration rate of ≥99% for Level 1 loads and ≥90% for Level 2 loads, and then maximizing the restoration rate of Level 3 loads.

[0051] The self-healing control method proposed in this invention achieves a fundamental reconstruction of the fault handling process in distribution networks through a tightly coupled, dynamically closed-loop intelligent decision-making system. This invention is the first to deeply integrate real-time topology generation, multi-objective optimization, dynamic simulation pre-verification, and execution feedback closed-loop into an organic whole with autonomous learning and adaptive capabilities. This method not only achieves dynamic and accurate topology perception through graph traversal algorithms, but more importantly, it constructs a fully automated closed-loop process of "perception-decision-pre-simulation-execution-verification-re-optimization." Specifically, the strategy output by the multi-objective optimization model needs to be dynamically simulated for safety verification. If the verification fails, the system automatically transforms the limit-crossing information exposed by the simulation into new constraints, feeding them back to the optimization model for iterative recalculation until a safe and feasible strategy is generated. In the execution phase, the system can also update the equipment status and re-optimize the strategy in real time based on the success or failure of the actual switch actions. This closed-loop control logic, driven by real-time data and possessing online safety pre-simulation and self-correction capabilities, effectively overcomes the shortcomings of traditional methods, such as the disconnect between optimization and safety, rigid strategies, and lack of adaptability, significantly improving the reliability, safety, and intelligence level of distribution network self-healing.

[0052] Example 2 This invention also discloses a line self-healing control system driven by real-time data of a distribution network, comprising a memory and a processor. The memory stores a line self-healing control method program driven by real-time data of the distribution network. When the processor executes the line self-healing control method program driven by real-time data of the distribution network, it performs the following steps: The data acquisition module collects real-time operational data of the power distribution network and uses a graph traversal algorithm to generate a real-time network topology. The fault diagnosis module monitors the electrical quantity data of key nodes in the real-time network topology and diagnoses faults based on this electrical quantity data. The fault location module combines the alarm status information of the fault indicator with the real-time network topology to locate the faulty section; The self-healing strategy generation module generates the optimal self-healing strategy for isolating faulty sections and restoring power supply to non-faulty sections based on a multi-objective optimization algorithm. The strategy security verification module performs dynamic simulation to verify the security of the optimal self-healing strategy. If the security check passes, the execution module will issue the optimal self-healing strategy for execution and verify the execution effect in real time.

[0053] It should be noted that by constructing unified, high-quality, and effective operational data and performing dynamic topology analysis, a realistic and reliable power grid state foundation is provided for fault location and self-healing strategy generation, significantly improving the accuracy of decision-making and response speed. Simultaneously, leveraging multi-objective optimization algorithms, under strict constraints on line load, a comprehensive balance is struck between multiple indicators such as the number of switching operations and power restoration rate. This overcomes the limitations of traditional methods that only pursue maximum power restoration, automatically generating strategies that achieve the optimal balance between operational risk, restoration efficiency, and operational economy. This upgrades self-healing control from simply "restoring power" to "safe, economical, and efficient optimal restoration." Furthermore, by introducing dynamic simulation safety verification before execution, risks such as line overload and voltage exceeding limits can be proactively identified and avoided, preventing secondary faults caused by blind operations and enhancing system safety and reliability. Combined with execution feedback and load priority management, the system possesses a high degree of intelligence and adaptability to cope with complex scenarios such as switch failure and limited power supply capacity, ultimately forming a complete technical closed loop of "perception-decision-verification-execution-feedback" to achieve intelligent self-healing control.

[0054] According to an embodiment of the present invention, the real-time acquisition of effective operation data of the distribution network and the generation of real-time network topology using a graph traversal algorithm include: Real-time acquisition of effective operation data of the power distribution network and standardization processing and data quality verification; Based on the electrical wiring diagram of the power distribution network, an initial global topology is constructed with power distribution equipment as vertices and electrical connections between equipment as edges; Real-time opening / closing status data of all switches are extracted from the standardized and quality-verified operating data and mapped onto the initial global topology to generate a dynamic topology. Using all power supply points as root nodes, a preset breadth-first search algorithm is used to traverse the dynamic topology, generating a real-time network topology containing information about each connected subgraph and its power supply path.

[0055] It should be noted that the raw operational data collected from the distribution terminals is cleaned and organized, and its completeness and rationality are verified. Abnormal data is identified and processed to ensure data quality and provide a reliable foundation for subsequent analysis. Then, various power distribution equipment (including circuit breakers, load switches, sectionalizing switches, fault indicators, transformers, etc.) are abstracted as vertices of a graph, and the electrical connection segments between equipment are abstracted as edges of the graph, generating an initial global topology. The real-time status of all switches (including circuit breakers, load switches, sectionalizing switches, etc.) is extracted from the preprocessed operational data and "mapped" onto the initial global topology (e.g., closing switches correspond to "edge connected," and opening switches correspond to "edge disconnected"), generating a "dynamic topology" that reflects the current actual connection relationships. The real-time network topology includes the current grid connectivity partitions and power supply path information: each connected subgraph represents an area independently powered by a single power source, and the power supply path information includes the electrical connection relationships and power flow direction from the power source point to each load node within the area.

[0056] According to an embodiment of the present invention, the real-time monitoring of electrical quantity data of key nodes in the real-time network topology, and the fault determination based on the electrical quantity data, includes: Real-time acquisition of three-phase current, zero-sequence current, and voltage of the power source root node in the real-time network topology; If any phase current exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting, and the duration exceeds the preset time threshold, and the voltage value is lower than the preset voltage threshold, then a fault is determined to have occurred in the distribution network.

[0057] It should be noted that, in order to achieve accurate fault identification and avoid misjudgment or omission, the power source root node is selected as the key monitoring node (because changes in the electrical quantities of the power source node can reflect the state of the entire connected sub-diagram). The three-phase current, zero-sequence current and voltage are collected in real time. When making a judgment, three conditions must be met simultaneously: the current of any phase exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting (covering overcurrent and grounding faults), the abnormal duration exceeds the preset time threshold (excluding transient interference), and the voltage value is lower than the preset voltage threshold (further confirming the fault rather than load fluctuation). If all three conditions are met, the distribution network is judged to have a fault.

[0058] According to an embodiment of the present invention, the step of locating the faulty section by combining the alarm status information of the fault indicator and the real-time network topology includes: Obtain the alarm status information of the fault indicator and its corresponding ID; Based on the fault indicator ID, its alarm status information is mapped to the corresponding position in the real-time network topology, and its associated connected subgraph is queried based on the position information. Based on the power supply path information corresponding to the connected subgraph, the alarm status of each fault indicator on the path is queried sequentially. The section between the last fault indicator that triggered an alarm and the first fault indicator that did not trigger an alarm is defined as the fault section.

[0059] It should be noted that, assuming that the system does indeed have a fault, the fault indicator ID is used to map the alarm status information (alarm or no alarm) to the corresponding location in the real-time network topology in order to accurately locate the smallest range (segment) where the fault occurred.

[0060] According to an embodiment of the present invention, the optimal self-healing strategy for isolating faulty sections and restoring power supply to non-faulty sections based on a multi-objective optimization algorithm includes: A multi-objective optimization model is constructed with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit. A pre-defined multi-objective genetic algorithm is used to solve the multi-objective optimization model, and a Pareto front solution set containing multiple candidate self-healing strategies is obtained. A comprehensive evaluation function is constructed based on the power restoration rate and the total number of switching operations. The comprehensive evaluation function is then used to calculate the comprehensive evaluation value corresponding to each candidate self-healing strategy. The candidate self-healing strategy with the highest comprehensive evaluation value is selected as the optimal self-healing strategy.

[0061] It should be noted that, by leveraging a multi-objective optimization algorithm, and under strict constraints on line load, this approach comprehensively weighs multiple indicators such as the number of switching operations and the power restoration rate. This overcomes the limitations of traditional methods that only pursue the maximum restored power volume. It can automatically generate a strategy that achieves the optimal balance between operational risk, restoration efficiency, and operational economy, upgrading self-healing control from simply "restoring power" to "safe, economical, and efficient optimal restoration." The power restoration rate refers to the ratio of the total load successfully restored after the self-healing strategy is executed to the total load of the non-faulty section that lost power due to the fault. In this embodiment, the comprehensive evaluation function is obtained by multiplying the power restoration rate and the total number of switching operations by weighting coefficients and then taking the difference. The weighting coefficients are dynamically determined based on the real-time operating scenario. Each candidate self-healing strategy includes a defined set of switching operations, the operation sequence, and the reconstructed network structure.

[0062] According to an embodiment of the present invention, the step of performing simulated security verification of the optimal self-healing strategy through dynamic simulation includes: Obtain the physical parameters of the distribution network, and construct a dynamic simulation model of the distribution network based on real-time operating data and the physical parameters of the distribution network; The reconstructed network structure data and switching operation sequence in the optimal self-healing strategy are input into the dynamic simulation model to obtain simulation results. The simulated electrical quantities are extracted from the simulation results, including the voltage values ​​of each node and the short-circuit current values ​​of each line. Calculate the deviation between the voltage value of each node and the preset rated voltage, and determine whether the deviation is less than the preset deviation threshold. Determine whether the short-circuit current value of each line is less than the rated breaking current value of its corresponding circuit breaker; If both judgments are yes, the verification passes and the optimal self-healing strategy is executed. Otherwise, the over-limit location is determined based on the simulation results, and the deviation value, short-circuit current value and their corresponding over-limit location data are input into a preset constraint condition lookup table to obtain the corresponding hard constraint conditions; The hard constraints are fed back to the multi-objective optimization model, and the corrected optimal self-healing strategy is generated by solving the problem again.

[0063] It should be noted that when performing dynamic simulation to verify the optimal self-healing strategy for safety, if the node voltage deviation exceeds the preset deviation threshold or the line short-circuit current exceeds the rated breaking current of the corresponding circuit breaker, it indicates that the self-healing strategy violates safety operation constraints. In this case, the simulation results need to be analyzed to identify the specific violations, and the hard constraints should be determined through a preset constraint reference table. For example, the simulation results may determine whether the short-circuit current of a certain line exceeds the rated breaking current of its terminal circuit breaker, or whether the deviation of the node voltage from the rated voltage in a certain area exceeds the allowable range. The hard constraints (such as "the short-circuit current of line A shall not exceed 12kA" or "the voltage deviation of node B shall be controlled within ±5%) are obtained by looking up the constraint reference table and directly fed back into the multi-objective optimization model. When the model is re-solved, these new constraints will be included in the optimization boundary along with the original line load rate constraints, thereby generating a modified optimal self-healing strategy that can avoid such safety risks, ensuring that the strategy meets both the power supply restoration objective and the requirements for safe operation of the distribution network.

[0064] According to an embodiment of the present invention, if the security verification passes, the optimal self-healing strategy is issued and executed, and the execution effect is verified in real time, including: Within a preset time period, determine whether a corresponding switch change confirmation signal has been received; if so, the execution is considered successful. If not, the switch that failed to execute is marked as inoperable, and this state is fed back to the multi-objective optimization model as a new constraint to regenerate the optimal self-healing strategy after excluding this switch.

[0065] It should be noted that if the safety check passes, the optimal self-healing strategy will be issued and executed, and the execution effect will be verified in real time. The focus is on solving the fault tolerance problem of switch operation failure. After the strategy is issued, the switch change confirmation signal will be monitored within a preset time period. If the signal is received, the operation is determined to be successful. If no signal is received, the switch is marked as inoperable. This state is fed back to the multi-objective optimization model as a new constraint, and the optimal self-healing strategy to exclude the switch is regenerated to avoid the interruption of the self-healing process due to a single switch failure.

[0066] Switches that fail to execute are marked as inoperable, and this state is fed back as a new constraint to the multi-objective optimization model, specifically including: Add the switch identifier that failed to execute to the set of inoperable switches; When resolving the multi-objective optimization model, the decision variable corresponding to the switch in the set is fixed to the current state and removed from the search space of the optimization variables; The search space is re-optimized to generate the optimal self-healing strategy after excluding all inoperable switches.

[0067] According to an embodiment of the present invention, it further includes: If the power supply recovery rate of the non-faulty section is less than the preset power supply recovery rate threshold after the optimal self-healing strategy is implemented; Obtain the predefined priority labels for each load node from the distribution network management system, including primary load, secondary load, and tertiary load; Based on the priority labels, set corresponding power recovery rate constraints for load nodes with different priorities; The minimum power restoration rate constraint is integrated as a new hard constraint into the multi-objective optimization model, and the modified optimal self-healing strategy is generated by solving the model again.

[0068] It should be noted that load priority labels are defined in the distribution network dispatching system, dividing all load nodes into three levels: Level 1 loads (critical loads): such as hospital intensive care units, city traffic lights, and emergency command centers, which require priority power restoration; Level 2 loads (important loads): such as residential communities and commercial complexes; Level 3 loads (general loads): such as industrial auxiliary equipment and non-essential commercial facilities. Priority is given to ensuring a restoration rate of ≥99% for Level 1 loads and ≥90% for Level 2 loads, and then maximizing the restoration rate of Level 3 loads.

[0069] This invention discloses a data-driven line self-healing control method and system for distribution networks. The data-driven line self-healing control method and system provided in this application provide a real and reliable power grid state basis for fault location and self-healing strategy generation by constructing unified and high-quality real-time operation data and performing dynamic topology analysis, which significantly improves the accuracy of decision-making and response speed. Meanwhile, leveraging multi-objective optimization algorithms, and under strict constraints on line load, the system comprehensively weighs multiple indicators such as the number of switch operations and power restoration rate. This overcomes the limitations of traditional methods that only pursue the maximum amount of restored power. It can automatically generate strategies that achieve the optimal balance between operational risk, restoration efficiency, and operational economy, upgrading self-healing control from simply "restoring power" to "safe, economical, and efficient optimal restoration." Furthermore, by introducing dynamic simulation safety verification before execution, it can proactively identify and avoid risks such as line overload and voltage exceeding limits, preventing secondary faults caused by blind operations and enhancing system safety and reliability. Combined with execution feedback and load priority management, the system possesses a high degree of intelligence and adaptability to cope with complex scenarios such as switch failure and limited power supply capacity. Ultimately, it forms a complete technical closed loop of "perception-decision-verification-execution-feedback," achieving intelligent self-healing control.

[0070] Example 3 The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of Embodiment 1.

[0071] Example 4 The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the embodiments.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A line self-healing control method driven by real-time data in a distribution network, characterized in that, include: Real-time acquisition of effective operational data of the power distribution network, and generation of real-time network topology based on the operational data; Real-time monitoring of electrical quantity data of the power root node in the real-time network topology, and fault determination based on the electrical quantity data; After determining that a fault has occurred, the faulty section is located by combining the alarm status information of the fault indicator and the real-time network topology. A multi-objective optimization model is constructed with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit. The optimal self-healing strategy for isolating the faulty section and restoring power supply to the non-faulty section is generated based on the multi-objective optimization model. The optimal self-healing strategy is simulated and its security is verified through dynamic simulation. If the security check passes, the optimal self-healing strategy will be deployed and executed, and the execution effect will be verified in real time.

2. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: The process of collecting real-time operational data of the power distribution network and generating a real-time network topology based on that data includes the following steps: The operational data is standardized and its quality is verified to obtain the processed operational data. Construct the electrical wiring diagram of the power distribution network, and based on the electrical wiring diagram, construct an initial global topology structure with power distribution equipment as vertices and electrical connections between equipment as edges; Extract the real-time opening / closing status data of all switches from the processed operating data, map it onto the initial global topology, and generate a dynamic topology. Using all power supply points as root nodes, the dynamic topology is traversed to generate a real-time network topology containing information on each connected subgraph and power supply path.

3. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: The real-time monitoring of electrical quantity data of the power root node in the real-time network topology, and the fault determination based on the electrical quantity data, specifically includes the following steps: The electrical quantity data includes the three-phase current, zero-sequence current, and voltage at the power source root node; If any phase current exceeds the preset overcurrent setting or the zero-sequence current exceeds the preset grounding setting, and the duration exceeds the preset time threshold, and the voltage value is lower than the preset voltage threshold, then a fault is determined to have occurred in the distribution network.

4. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: The specific steps for locating the faulty segment by combining the alarm status information of the fault indicator and the real-time network topology include: Obtain the alarm status information and corresponding ID of the fault indicator; Based on the ID, the alarm status information is mapped to the corresponding position in the real-time network topology, and the connected subgraph to which it belongs is queried according to the information of the position; Based on the power supply path information corresponding to the connected subgraph, the alarm status of each fault indicator on the path is queried sequentially. The section between the last fault indicator that triggered an alarm and the first fault indicator that did not trigger an alarm is defined as the fault section.

5. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: The method for generating the optimal self-healing strategy for isolating the faulty section and restoring power supply to the non-faulty section based on the multi-objective optimization algorithm includes: Solving the multi-objective optimization model yields a Pareto front solution set containing multiple candidate self-healing strategies; A comprehensive evaluation function is constructed based on the power restoration rate and the total number of switching operations. The comprehensive evaluation function is then used to calculate the comprehensive evaluation value corresponding to each candidate self-healing strategy. The candidate self-healing strategy with the highest comprehensive evaluation value is selected as the optimal self-healing strategy.

6. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: The specific steps for simulating and verifying the optimal self-healing strategy through dynamic simulation include: Obtain the physical parameters of the distribution network, and construct a dynamic simulation model of the distribution network based on the processed operating data and the physical parameters of the distribution network. The reconstructed network structure data and switching operation sequence in the optimal self-healing strategy are input into the dynamic simulation model to obtain simulation results. The simulated electrical quantities are extracted from the simulation results, including the voltage values ​​of each node and the short-circuit current values ​​of each line; Calculate the deviation between the voltage value of each node and the preset rated voltage, and determine whether the deviation is less than the preset deviation threshold. Determine whether the short-circuit current value of each line is less than the rated breaking current value of its corresponding circuit breaker; If both judgments are yes, the verification passes and the optimal self-healing strategy is executed. Otherwise, the over-limit location is determined based on the simulation results, and the deviation value, short-circuit current value and their corresponding over-limit location data are input into a preset constraint condition lookup table to obtain the corresponding hard constraint conditions; The hard constraints are fed back to the multi-objective optimization model, and the corrected optimal self-healing strategy is generated by solving the problem again.

7. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: If the security check passes, the optimal self-healing strategy will be deployed and executed, and the execution effect will be verified in real time. The specific steps include: Within a preset time period, determine whether a corresponding switch change confirmation signal has been received; if so, the execution is considered successful. If not, the switch that failed to execute is marked as inoperable, and the inoperable state is fed back to the multi-objective optimization model as a new constraint to regenerate the optimal self-healing strategy after excluding this switch.

8. The line self-healing control method driven by real-time data of a distribution network according to claim 7, characterized in that: The steps of marking the failed switch as inoperable and feeding the inoperable state back to the multi-objective optimization model as a new constraint include: Add the switch identifier that failed to execute to the set of inoperable switches; When resolving the multi-objective optimization model, the decision variable corresponding to the switch in the set is fixed to the current state and removed from the search space of the optimization variables; The search space is re-optimized to generate the optimal self-healing strategy after excluding all inoperable switches.

9. The line self-healing control method driven by real-time data of a distribution network according to claim 1, characterized in that: Also includes: If, after implementing the optimal self-healing strategy, the power supply recovery rate of the non-faulty section is less than the preset power supply recovery rate threshold; Then obtain the predefined priority labels of each load node, including primary load, secondary load and tertiary load; Based on the priority labels, set corresponding power recovery rate constraints for load nodes with different priorities; The minimum power recovery rate constraint is integrated as a new hard constraint into the multi-objective optimization model, and the modified optimal self-healing strategy is generated by solving the model again.

10. A line self-healing control system driven by real-time data of a distribution network using the method described in any one of claims 1-9, comprising a data acquisition module, a fault judgment module, a fault location module, a self-healing strategy generation module, a strategy security verification module, and an execution module, characterized in that: The data acquisition module collects real-time operational data of the power distribution network and generates a real-time network topology based on the operational data. The fault diagnosis module monitors the electrical quantity data of the power root node in the real-time network topology and diagnoses faults based on the electrical quantity data. The fault location module, after determining that a fault has occurred, locates the faulty segment by combining the alarm status information of the fault indicator and the real-time network topology. The self-healing strategy generation module constructs a multi-objective optimization model with the objective function of minimizing the number of switching operations and maximizing the power supply recovery rate of non-faulty sections, and with the hard constraint that the line load rate does not exceed the safety limit; based on the multi-objective optimization model, it generates the optimal self-healing strategy for isolating the faulty section and restoring the power supply to the non-faulty section. The strategy security verification module performs simulated security verification on the optimal self-healing strategy through dynamic simulation. If the security check passes, the execution module will issue the optimal self-healing strategy for execution and verify the execution effect in real time.

11. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-9.

Images